asciimonster {l Wrote}:Could somebody explain what properties are assigned to end camera's?
I can't seem to understand the explanation on
this page
Could you perhaps try to improve the docs (once you understand what I meant)?
This is as far as I came:
- type = ignore -> Don't use as end cam
This is a general property for all objects: anything with type=ignore will be ignored, so that's not camera specific.
- type = "" -> Use as end cam (but that is not the behaviour I have seen)
This is the default, and atm the same as 'ahead'. Though that might obviously change in the future, so best is to define one of the currently supported types.
- type = "ahead" -> Seems to be the default now
- type = "fixed" -> What's the difference with a single end camera?
Ahead is the default camera - the one you would get if you don't specify anything for any camera. The camera flies ahead of the kart, showing the end animation.
The 'fixed' camera is a camera that is at a fixed position. It points at the kart and zooms in so that the kart is nicely shown. Try the island track to see how this camera works. The position of the camera in blender is used to place the camera, while an 'ahead' camera does not use the position the camera is at directly (but see below).
- start = ??? -> I have no idea what type this property should be...
One thing that needs to be defined is when to use what camera. First of all the cameras are sorted depending on the quads of the driveline they are closest to (that sounds complicated - it basically means that the first camera is the one closest to the start line, the next camera is the one the kart reaches next when it keeps on driving along the driveline). This defines in which order the cameras will be used. When a kart crosses the lap line, the first camera in this order is activated (independent of any other settings). If the camera is an 'ahead of kart camera', it means that the camera will be moved (from the position in blender) to just ahead of the kart; if it's a fixed camera this camera will turn torwards the kart and zoom in.
The actual time when the switch to the next camera is happening depends on the 'start' value: imagine a sphere around each camera position in blender. After the first end camera is activated as described above, the kart will keep on driving. The moment it crosses the sphere of the next camera, the next camera will become the active camera (it 'starts' - that's why I called it start, though perhaps a better name might be found). Again, if that camera is an 'ahead of kart' camera, it will be positioned just ahead of the kart, and if it's a fixed camera, it will turn towards the kart and zoom in.
So the location together with the 'start' value allow you to define
when to switch on the camera.
I hope that's a bit clearer?
I think it's easier to set type="endcamera" to use it as an end cam rather than type being empty. Then a second property "endcamtype" defines it's behaviour.
I don't recommend to let the type empty for a camera (perhaps the track exporter should print a warning and ignore the camera??). Your approach needs to define two properties, while mine only needs one (basically type=endcamera is deduced from the fact that the object is a camera, and your endcamtype becomes my type).
P.S. Are particle-emitters and billboards attached to objects, and if so which ones?
I am currently re-working the billboards: they will be attached to meshes with 4 vertices and one image (this is already the case, but the whole code to export it will be rewritten, esp. it will allow billboards to be IPO animated). I haven't really looked at the particle-emitter code.Ideally we would use the data from blender's particle emitter somehow (so that a blender rendering would be at least similar to the stk version), but if this is too complicated, we will probably just attach it to a mesh with some properties - samuncle, any comments here?
Cheers,
Joerg